Practical Guidance for Measuring Handwashing … Guidance for Measuring Handwashing Behavior: 2013...

Practical Guidance for Measuring Handwashing Behavior: 2013 Update

Pavani Ram, MD

February 2013

The Water and Sanitation Program is a multi-donor partnership administered by the World Bank to support poor people in obtaining affordable, safe, and sustainable access to water and sanitation services.

Global Scaling Up Handwashing

WATER AND SANITATION PROGRAM: WORKING PAPER

8714-Book.pdf i8714-Book.pdf i 2/8/13 10:06 AM2/8/13 10:06 AM

By Pavani Ram, MDUniversity at Buffalo, The State University of New York

The first edition of “Practical Guidance for Measuring Handwashing Behavior” was published by the Water and Sanitation

Program in 2010. There has been substantial research relevant to handwashing behavior measurement since the previous

publication. Based on the substantial continued interest in measuring handwashing behavior among researchers and

practitioners alike, we present here the first update to this document. We have updated the format to address the validity

of each measure as compared with other handwashing measures and health outcomes, potential for bias or data collection

errors, use in evaluating handwashing programs, as well as the bottom line for researchers and practitioners.

Christina Crabtree Ide, Kelly Kamm, Jelena Vujcic, and Anne Weaver contributed extensively to the review of the recent

handwashing literature. Gratitude goes to Benjamin Arnold, Bertha Briceño, Claire Chase, Craig Kullmann, and Amy Pickering

for thoughtful reviews of this document. The following individuals, in alphabetical order, contributed to the thinking presented

in the first edition of this guidance document: Adam Biran, Anna Bowen, Val Curtis, Jacqueline Devine, Stewart Granger,

Orlando Hernandez, Steve Luby, Jack Molyneaux, Eddy Perez, and Wolf-Peter Schmidt. Sincere thanks go to them for their

constant interest in this topic and their intellectual generosity.

Global Scaling Up Handwashing is a project by the Water and Sanitation Program (WSP) focused on applying innovative

behavior change approaches to improve handwashing with soap behavior among women of reproductive age (ages 15–49)

and primary school-age children (ages 5–9). It was implemented by local and national governments with technical support

from WSP in four countries: Peru, Senegal, Tanzania, and Vietnam. For more information, please visit www.wsp.org/

scalinguphandwashing.

This Working Paper is one in a series of knowledge products designed to showcase project findings, assessments,

and lessons learned in the Global Scaling Up Handwashing project. This paper is conceived as a work in progress to

encourage the exchange of ideas about development issues. For more information, please email Pavani Ram at

[email protected] or visit www.wsp.org.

WSP is a multi-donor partnership created in 1978 and administered by the World Bank to support poor people in obtaining affordable, safe, and sustainable access to water and sanitation services. WSP’s donors include Australia, Austria, Canada, Denmark, Finland, France, the Bill & Melinda Gates Foundation, Ireland, Luxembourg, Netherlands, Norway, Sweden, Switzerland, United Kingdom, United States, and the World Bank.

WSP reports are published to communicate the results of WSP’s work to the development community. Some sources cited may be informal documents that are not readily available.

The findings, interpretations, and conclusions expressed herein are entirely those of the author and should not be attributed to the World Bank or its affiliated organizations, or to members of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work.

The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to [email protected]. WSP encourages the dissemination of its work and will normally grant permission promptly. For more information, please visit www.wsp.org.

© 2013 Water and Sanitation Program

8714-Book.pdf ii8714-Book.pdf ii 2/8/13 10:06 AM2/8/13 10:06 AM

www.wsp.org/scalinguphandwashing

www.wsp.org iii

Executive Summary

ogy may eventually be useful for measuring household en-vironmental contamination.

Structured observations have been used frequently in the handwashing literature. These observations require trained observers to watch and record household handwashing and related behaviors, and yield details about handwashing at critical times, such as after defecation. Handwashing behav-ior, as recorded during structured observation, has been asso-ciated with disease risk. However, individuals are reactive to the presence of an outside observer. While the reactivity gen-erated by structured observation warrants caution, structured observations remain a cornerstone of handwashing measure-ment because of the rich details yielded by them.

Sensors have been used in several studies to provide an ob-jective record of the number and timing of soap-use events. A variety of sensors have been tested in research studies in-cluding accelerometers embedded in bars of soap, infrared sensors at entries to restrooms coupled with monitoring of liquid soap use, and radiofrequency-controlled transmitters worn by subjects with readers at handwashing locations. The use of sensor technology is promising in select settings, despite facing several challenges: These methods are rela-tively expensive, because of specialized hardware and per-sonnel costs for analysis of sensor-elicited data. In addition, some of the sensor methods do not generate respondent-specific information and some do not inform about rates of handwashing with soap at critical times, such as after def-ecation. In spite of these caveats, sensor-based handwash-ing measurement yields objectivity and reliability and, thus, further evaluation is clearly warranted.

Based on these assessments, the following recommenda-tions are made for various types of studies carried out in low- and middle-income countries, including well-funded

projects, projects with minimal funding, and mixed-purpose, large population-based surveys.

For well-funded projects, the most rigorous methods should be used. Structured observations and rapid observations

In low- and middle-income settings, accurate measures of handwashing behavior are critical to understanding house-holds’ health environment. But it can be challenging to measure handwashing. This document discusses a set of handwashing indicators and recommendations prepared to support the Water and Sanitation Program’s Global Scal-ing Up Handwashing project carried out in four countries. Descriptions of these measures, and the recommendations for their use (Table 1, page 20), should be of interest to a broad audience.

The following handwashing measures are assessed based on their validity, reliability, and efficiency:

Self-reports via questionnaire are the easiest way to measure handwashing. Several studies have shown a relationship be-tween self-reported handwashing behavior and disease risk. But, individuals often report better handwashing behavior than they display during observation. This exaggeration of true behavior may result from a perceived high social desir-ability of handwashing. Questionnaires remain an impor-tant source of information about handwashing knowledge and other determinants of handwashing behavior.

Rapid observations include several easily collected valid and reliable indicators. These include observations on the availability of soap and water, the presence of these tools at dedicated handwashing stations, and inspections of hand cleanliness. While these indicators do not directly indicate handwashing behavior, they are currently used as surrogate markers because they are reliable and efficient. But evidence of how well they predict actual handwashing behavior and disease risk is still forthcoming.

Microbiological measures of hand contamination are ob-jective measures of hand contamination, and consequently would seem desirable. However, this is currently a costly way to assess hand cleanliness. Large sample sizes may be needed to overcome the variability noted in results of hand microbiology testing. If the cost of such measurement can be decreased and reliability improved, hand microbiol-

8714-Book.pdf iii8714-Book.pdf iii 2/8/13 10:06 AM2/8/13 10:06 AM

iv Global Scaling Up Handwashing

efficient source of household-level handwashing informa-tion. Rapid observations are markers for actual behavior. Self-reports may be used to measure knowledge and other possible determinants of handwashing behavior. Minimally funded studies that need affordable yet reliable methods to monitor handwashing behavior may warrant an investment in sample size estimates by a statistician or epidemiologist. These investments can frequently pay for themselves, as sample needs are frequently much lower than expected.

For mixed-purpose, large population surveys, such as the Demographic and Health Survey (DHS) or the Multiple Indicator Cluster Survey (MICS), where handwashing is only one of many behaviors of interest, rapid observations

are recommended as the most efficient method of measur-ing handwashing behavior.

should be used to obtain objective measures of handwash-ing. It is assumed that well-funded studies have the resources to involve experienced researchers with research and statisti-cal expertise. Self-reported measures may be used to learn about knowledge and other determinants of handwashing. Hand contamination and sensor-based measures should be considered, as much is still to be learned of them. These well-funded studies should also continue to measure health outcomes to better document the relationship between hand-washing promotion and health outcomes, as well as between measured handwashing behaviors and health outcomes.

Studies with minimal funding should consider carrying out structured observations in a small sample of households, primarily to assess change in behaviors targeted by the handwashing intervention. Rapid observations are the most

Practical Guidance for Measuring Handwashing Behavior: 2013 Update Executive Summary

8714-Book.pdf iv8714-Book.pdf iv 2/8/13 10:06 AM2/8/13 10:06 AM

www.wsp.org v

Table of Contents Executive Summary ...................................................................iii I. Introduction ............................................................................... 1 II. Methods of Measuring Handwashing Behavior ....................... 3 Self-report .................................................................................. 3 Proxy Measures: Measurement of Microbiological Hand Contamination ............................................................... 5 Proxy Measures: Rapid Observation of Handwashing Materials ................................................................................. 7 Proxy Measures: Hand Cleansing Product Consumption .......... 8 Proxy Measures: Observation of Behavior During Handwashing Demonstration ................................................. 9 Proxy Measures: Visual Inspection of Hand Cleanliness ......... 10 Direct Measures: Structured Observations .............................. 11 Direct Measures: Video Observation ........................................ 14 Direct Measures: Sensors ........................................................ 15 Use of Composite Measures .................................................... 17 III. Conclusion ............................................................................... 19 IV. Recommendations for Various Scenarios .............................. 23 Well-funded Handwashing Promotion Programs or Research Studies ............................................................. 23 Handwashing Promotion Programs with Minimal Funding ...... 23 Nationally Representative Surveys, e.g., DHS or MICS Surveys ................................................................... 24 References ............................................................................... 25

Table 1: Summary of Strategies to Measure Handwashing

Behavior ............................................................................. 20

8714-Book.pdf v8714-Book.pdf v 2/8/13 10:06 AM2/8/13 10:06 AM

8714-Book.pdf vi8714-Book.pdf vi 2/8/13 10:06 AM2/8/13 10:06 AM

www.wsp.org 1

Advocates of promoting handwashing with soap agree that this behavior has important health benefits across the globe, especially in low- and middle-income countries, including for the prevention of diarrhea and acute respi-ratory infections, such as influenza.1 Efforts, both large-scale and focused, are underway worldwide to pro-mote handwashing with soap at the community level (www.globalhandwashing.org). Although there is broad agreement about the health benefits of promoting hand-washing with soap, there is not similar agreement about the best ways to measure the behavior that these promo-tion programs set out to change. Most intervention stud-ies report health outcomes data but they do not provide information about effects of the program on handwashing behavior.2 Vindigni and colleagues noted that, as of their writing, only 27 unique studies carried out in low- and middle-income countries had evaluated behavior change resulting from handwashing promotion.3 Among these 27 studies, proxy measures and self-reported behavior were commonly used, with only rare use of structured observa-tions to directly measure handwashing behavior. There is a pressing need to appreciate the limitations and challenges to these and other methods of measuring handwashing be-havior, as well as to develop a pragmatic route forward for the evaluation of handwashing promotion programs, which are increasingly being implemented in dozens of countries worldwide.

The reality is this: there is no universally applicable method for measuring handwashing behavior that is valid, relevant, affordable, and logistically feasible for the various settings in which such behavior might need to be measured. The aim of this document is to describe tech-niques and to propose strategies for measuring hand-washing behavior in low- and middle-income country contexts.

IntroductionI.Measures of handwashing behavior may be scrutinized with respect to the following criteria:

• Validity: “an expression of the degree to which a mea-surement measures what it purports to measure”4

The validity of a measure of handwashing may be evaluated based on how it compares to other mea-sures of handwashing. An individual who demon-strates good handwashing practice, as indicated by one measure, may be expected to demonstrate good handwashing practice, as indicated by other mea-sures. For example, self-reported handwashing may be compared to directly observed handwashing be-havior, as measured by structured observation, or compared to the presence of handwashing materials at designated locations.

Since improved health is the ultimate goal of hand-washing, the validity of a handwashing measure may also be assessed by comparing to health outcomes. For example, the disease experience of those who report washing hands may be compared to the dis-ease experience of those who do not report washing hands. Evaluating a handwashing measure against a health outcome, such as diarrhea, provides informa-tion on whether the handwashing measure is rele-vant to population health.

• Efficiency: “the effects or end results achieved in relation to the effort expended in terms of money, resources, and time”5

All measurement of handwashing is challenged by the complexities of this human behavior. An individual may wash hands with soap in the context of some critical times for pathogen acquisition or transmission, such as after

1 Curtis 2003; Curtis and Cairncross 2003; Luby et al. 2004; Luby et al. 2005; Rabie and Curtis 2006; Aiello et al. 2008; Talaat et al. 2011.2 Cairncross et al. 2010.3 Vindigni et al. 2011.4 Last 2001.5 Ibid.

8714-Book.pdf 18714-Book.pdf 1 2/8/13 10:06 AM2/8/13 10:06 AM

Practical Guidance for Measuring Handwashing Behavior: 2013 Update Introduction

2 Global Scaling Up Handwashing

handwashing behavior. Self-reported, proxy, and directly observed measures are described. For each type of measure, we present information as available on the validity of the measure (as compared to other handwashing measures, as well as to health outcomes, where data are available), the efficiency of the measure, the potential for bias or data col-lection errors, the use of the measure in the evaluation of handwashing programs, other useful information for the researcher or practitioner, and the bottom line of the utility of the measure. We conclude with recommendations for the measurement of handwashing behavior in a variety of con-texts, including research and large nationally-representative surveys.

defecation, but not at others, such as before feeding a child. Thus, summarizing an individual’s overall handwashing be-havior requires taking into account variations in behavior at different critical times. Moreover, an individual may be in-consistent in her behavior, for example washing hands with soap after some defecation events but not all; such variation in reliability also risks the possibility of misclassifying an individual as “handwasher” or “non-handwasher.” Further-more, both reported and observed markers of handwashing behavior have been found to be significantly associated with socioeconomic status, making adjusting for this important explanatory factor extremely important.6

Described below are the positive and negative attributes of various commonly applied and novel methods of measuring

6 Luby and Halder 2008.


www.wsp.org 3

7 Pickering, Davis et al. 2010.8 Pickering, Boehm et al. 2010.9 Rhee et al. 2008.10 Luby and Halder 2008; Schmidt et al. 2009; Ram, Halder et al. 2010.11 Unicomb et al. 2010.12 Silk et al. 2010.13 Luby et al., Using Child Health Outcomes, 2011.

Self-reportThe easiest way to measure handwashing behavior is to use a questionnaire to ask the respondent directly about her behavior. The respondent may be questioned about how she washes hands (e.g., with or without soap), how recently she washed, how often she washes (e.g., on the day before interview), and when she washes. Information about hand-washing at critical times may be posed in several ways:

• an open-ended fashion: e.g., when do you wash your hands with soap

• in a closed-ended fashion: e.g., do you wash your hands with soap before feeding your child, or

• in a scaled fashion: e.g., how frequently do you wash your hands with soap before feeding your child? always, almost always, sometimes, or never?

ValidityComparison with other handwashing measures:In a pair of studies by Pickering and colleagues, self-reported handwashing behavior was not associated with hand contamination levels in one study7 but was found to be associated in the other study.8 Report of recent handwash-ing (�1 hour before specimen collection) was associated with lower levels of hand contamination, and individuals reporting always washing hands with soap after defecation had lower levels of E. coli and fecal streptococci than indi-viduals reporting sometimes or never washing hands with soap after defecation.

In our analysis of the endline data from the Impact Evalu-ation of the Global Scaling Up Handwashing project, in which we accounted for the frequency of reported soap use, we found moderate or greater agreement between self-reported handwashing after fecal contact and obser-vation of soap anywhere in the home in both Peru and Vietnam. In Vietnam, we found that persons reporting

handwashing with soap after fecal contact were more likely to be observed washing hands with soap after fecal contact (RRadj�2.93, 95% CI�1.53 � 5.64); we did not find similar relationships in Peru and Senegal (Ram, WSP paper in preparation).

Comparison with health outcomes:Several studies have found that groups with high self-reported handwashing behavior have lower disease risk than groups with relatively lower self-reported handwashing behavior. Two observational studies have reported asso-ciations between self-reported handwashing behavior and child mortality. Rhee and colleagues showed that mothers’ reports that the birth attendant washed her hands with soap and water before assisting with the delivery, and reports that she herself sometimes or always washed hands before handling the neonate, were associated with significantly reduced risk of neonatal mortality: maternal handwash-ing (RR 0.40, 95% CI 0.28 � 0.58) and birth attendant handwashing (RR 0.81, 95% CI 0.68 � 0.98).9 Unfortu-nately, the multivariate analysis yielding these risk ratios did not adequately account for socioeconomic status covari-ates, which are often shown to be strongly correlated with handwashing behaviors.10 Unicomb and colleagues have described a case-control analysis in which caregivers of chil-dren who died from diarrhea reported less frequent hand-washing (18 times per day) than caregivers of well children (21 times per day) (ORadj 0.90, 95% CI 0.89 � 0.95, in a multivariate model adjusting for wealth and education).11 In another case-control analysis, Silk and colleagues noted that caregivers of well children were more likely to report frequent handwashing (more than 12 times in the previous day) than caregivers of children with pneumonia.12 Most recently, Luby and colleagues published a study of nearly 500 households, in which maternal self-report of hand-washing with soap before feeding a child was associated with decreased diarrhea in the child.13

Methods of MeasuringHandwashing BehaviorII.


Practical Guidance for Measuring Handwashing Behavior: 2013 Update Methods of Measuring Handwashing Behavior


a handwashing promotion and soap provision interven-tion in India.17 Huda and colleagues describe that a pilot intervention was tested using self-reported handwashing as the principal behavioral outcome.18 Since self-reported handwashing was shown to increase in response to the in-tervention, an at-scale water, sanitation, and handwashing intervention was subsequently implemented in Bangladesh. However, a robust evaluation of the at-scale intervention using observational methods, and not self-report, demon-strated no behavior change attributable to the intervention. In Thailand, a handwashing and face mask intervention to prevent household transmission of influenza found that re-spondents in the intervention arm reported handwashing more frequently than in the control arm.19

Other useful information The large discrepancies in reported compared to observed behavior indicate that self-report is a limited measure of an individual’s true handwashing behavior. Questionnaires may also be used to elicit information relevant to behavioral determinants that may facilitate or impede handwashing; such determinants may include attitudes and beliefs, and logistical factors such as access to adequate quantities of water. One set of determinants that is commonly measured is knowledge related to handwashing behavior. Knowledge is the understanding of how, when, and why to wash hands. In contrast, behavior is the execution, i.e., the washing of hands with soap after one defecates. Questions regarding whether and when hands should be washed with soap may provide information about the respondent’s knowledge of appropriate handwashing behavior. However, knowledge does not equate to behavior, in that an individual may know well when to wash hands but the individual does not often practice handwashing at those times because of various barriers. Describing changes in knowledge of appropriate handwashing behavior, e.g., naming of critical times, may be useful as part of monitoring a handwashing promotion campaign that proposes to increase knowledge of hand-washing in the target population.

EfficiencyCollecting handwashing behavior information by self-report is efficient, since the data can be gathered quickly using questionnaires, among a large number of house-holds, at relatively low cost. Because of this efficiency, self-report is frequently included in assessments of hand-washing behavior. However, as outlined below, substantial concerns about bias suggest that sole reliance on self-report typically yield overly optimistic estimates of handwashing behavior.

Potential for bias or data collection errorsAwareness of the social desirability of handwashing may result in an individual’s overestimation of self-reported handwashing behavior. This overestimation has been dem-onstrated repeatedly, when self-reported behavior has been compared to observed behavior.14 In Bangladesh, whereas 77 percent of respondents reported washing hands with soap or ash after defecation, only 32 percent were observed to do so.15 Stanton and colleagues, and Biran and colleagues have each shown that there is poor agreement between reported behavior and observed behavior; however, these studies did not account for the high frequency of reported soap use, which can lead to an artificially low estimations of agree-ment.16 In our own recent analysis of endline data from the Impact Evaluation of the Global Scaling Up Handwashing project, we found that handwashing with soap after def-ecation was reported by 65 percent of caregivers in Peru, 45 percent of caregivers in Senegal, and 45 percent of care-givers in Vietnam (Ram, WSP paper in preparation). In contrast, soap was used for handwashing at only 34 percent, 25 percent, and 24 percent of fecal contact events in Peru, Senegal, and Vietnam, respectively.

Use in evaluation of handwashing programsSelf-reported handwashing behavior is commonly recorded in handwashing promotion program evaluations. For ex-ample, Biran and colleagues showed minimal change, and relatively rare reporting of soap use, before and after

14 Stanton et al. 1987; Manun’Ebo et al. 1997; Biran et al. 2008; Danquah 2010.15 “Handwashing Behavior in Rural Bangladesh.” 2008. Health and Science Bulletin Vol. 6, No. 3.16 Byrt et al. 1993; Sim and Wright 2005.17 Biran et al. 2009.18 Huda et al. 2012.19 Simmerman et al. 2011.



www.wsp.org 5

publications.22 Broadly, hand contamination data is re-ported as the number of colony-forming units (a micro-biological term used to denote the density of organisms) of the organism of interest per hand, or per volume of media in which hands were rinsed; organisms typically sought are fecal coliforms, E. coli (a subset of fecal coliforms), and fecal streptococci. Numerous studies have tested contamination of hands to compare the microbiological efficacy of differ-ent hand cleansing regimens.23 A number of studies have found that hand contamination is reduced after handwash-ing with soap, or after cleansing with waterless sanitizers.24

ValidityComparison with other handwashing measures:In our work in Bangladesh, we have found that hand con-tamination was neither associated with soap use observed during structured observation, nor correlated with the num-ber of times soap was used per day, as measured by accelerom-eters embedded in soap.25 Presence of soap in the household was not associated with levels of hand contamination in two studies carried out in Tanzania.26 However, visible dirt on the palm, finger pads, or under nails was significantly and in-versely associated with fecal streptococci and E. coli.27

Comparison with health outcomes:Evidence from Pakistan28 and Tanzania29 supports a positive as-sociation between hand contamination and health outcomes, meaning that children of mothers whose hands are more con-taminated have been found to have higher rates of diarrhea than children of mothers whose hands are less contaminated. Pickering and colleagues found in Tanzania that mean levels of fecal streptococci were positively associated with the prevalence of respiratory symptoms,30 but found an inverse relationship between the levels of E. coli on hands and respiratory symp-toms, a surprising finding that merits further exploration.31

Other possible determinants of handwashing behavior that can be measured by questionnaires relate to the opportunity to access handwashing tools (e.g., as in the FOAM framework—Focus on Opportunity, Ability, and Motivation—access to soap and water near a latrine), ability (e.g., capacity to ensure access to steady supply of soap), and motivation (e.g., beliefs about the importance of soap).20 Appropriate measurement of these and other possible determinants can be maximized by re-liance on a clearly considered framework for handwashing be-havior change.21 Since this document focuses on measurement of handwashing behavior, a review of frameworks to promote handwashing behavior change is beyond its scope.

The bottom lineAlthough there is some data to suggest that self-reported handwashing, in general or at specific critical times, is as-sociated with improved health, there is overwhelming evi-dence indicating that individuals overestimate their own handwashing behavior. Therefore, we do not recommend the use of self-reported handwashing behavior. Question-naires may be used, instead, to capture knowledge of critical times to wash hands, and to capture psychosocial con-structs, such as self-efficacy and social norms, which may influence handwashing behavior.

Proxy Measures: Measurement of Microbiological Hand ContaminationMeasurement of microbiological contamination of hands is another proxy measure of handwashing behavior. The un-derlying assumption is that hands that are washed with soap will be less contaminated with fecal organisms than hands that are not washed with soap. The details of measuring hand contamination, e.g., by fingertip rinses or hand imprints on semi-solid media, among others, are beyond the scope of this paper but are covered in numerous peer-reviewed

20 Coombes and Devine 2010.21 Hernandez et al. 2012.22 Pinfold 1990; Hoque et al. 1995; Luby, Agboatwalla et al. 2001; Luby, Agboatwalla et al. 2007; Judah et al. 2009; Pickering, Boehm et al. 2010; Pickering, Davis et al. 2010;

Burton et al. 2011.23 CDC 2002; Grayson et al. 2009; Luby, Kadir et al. 2010; Pickering, Boehm et al. 2010; Burton et al. 2011; Pickering, Davis et al. 2011.24 Pinfold 1990; Hoque et al. 1995; Luby, Agboatwalla et al. 2001; CDC 2002; Pickering, Boehm et al. 2010.25 Ram, Luby et al. 2010.26 Pickering, Boehm et al. 2010; Pickering, Davis et al. 2010.27 Pickering, Davis et al. 2010.28 Luby, Agboatwalla et al. 2007.29 Pickering, Davis et al. 2010.30 Pickering, Boehm et al. 2010; Pickering, Davis et al. 2010.31 Pickering, Davis et al. 2010.




different levels of contamination, since left hands may have more fecal contact than right hands.37

Use in evaluation of handwashing programsA few studies have used hand contamination to evaluate handwashing programs. In Thailand, Pinfold found that re-spondents in an intervention group had significantly lower finger contamination after the intervention, compared to the control group.38 In contrast, Luby and colleagues found no difference in hand contamination among mothers in an intervention group exposed to handwashing promotion and soap provision, and mothers in a comparison group that received no promotion or soap provision.39

Other useful informationContamination of mothers’ and children’s hands with fecal in-dicator bacteria was correlated with contamination of house-hold stored water.40 Hand contamination levels were associated with several household activities, including leaving the house-hold compound, bathing, preparing food, and eating, as mea-sured by structured observation in a study of 22 respondents.41

In a high-income country, evaluating the sheer presence or absence of organisms on hands, rather than the numbers of organisms, may be useful to predict prior handwashing behav-ior. In household environments lacking access to improved sanitation, far greater fecal contamination may be expected. Thus, the presence or absence approach for evaluating micro-bial hand contamination is unlikely to demonstrate sufficient heterogeneity in order to discriminate between “better” and “worse” handwashers; i.e., we anticipate that the vast major-ity of persons residing in households in low-income contexts would be found to have detectable organisms of fecal con-tact, irrespective of handwashing or other hygiene behaviors.

Organism-based testing (e.g., for E. coli or Rotavirus) may be feasible in study settings as molecular methods become less expensive.42

EfficiencyCurrently, measuring hand contamination is relatively expensive; in Bangladesh, laboratory-based microbiological testing for fecal coliforms and E. coli costs approximately US$10 per individual, well beyond the means of most program monitoring and evalu-ation budgets. In other sites, the cost of hand rinse sampling is US$2–3, which may be more approachable for modest budgets. A field-friendly method of measuring fecal coliforms and E. coli on hands is being sought and may prove to be an inexpensive and feasible method of testing for hand contamination.32

Potential for bias or data collection errorsHand contamination provides greater objectivity than self-report. But, there are limitations to the data resulting from microbiological testing of hands. First, handwashing does not necessarily eliminate all organisms from hands.33 Also, the level of hand contamination on an individual’s hands can vary greatly within the course of several hours; in Bangladesh, we found about three orders of magnitude differences between counts of fecal coliforms estimated at random and at critical times. In addition, there was rapid recontamination of hands in this environment, with fecal coliforms on hands of 100 percent, and E. coli on 80 per-cent of 25 participating women two hours after a thorough handwashing with soap.34 Thus, in a heavily fecally con-taminated environment, a single hand contamination mea-sure likely only reflects very recent handwashing behavior and not typical behavior of the individual. Time elapsed since hands were last washed with soap has been associ-ated with the level of hand contamination.35 Other factors affect hand contamination, and thus, the detection of or-ganisms on hands, including the presence of a wristwatch, jewelry, length of fingernails, and recent application of hand lotion.36 Duration since last fecal contact, and overall fecal contamination of the environment likely also impact the level of contamination detected on a subject’s hands. Also, in some cultures, an individual’s two hands may have

32 Wang et al. 2011.33 Pickering, Boehm et al. 2010.34 Ram, Jahid et al. 2011.35 Pickering, Julian et al. 2011.36 Fagernes and Lingaas 2011.37 Hoque et al. 1995.38 Pinfold and Horan 1996.39 Luby, Agboatwalla et al. 2007.40 Pickering, Davis et al. 2010.41 Pickering, Julian et al. 2011.42 Ibid.



www.wsp.org 7

we found similar associations in crude analyses but, after adjusting for wealth, the associations were not significant (Ram, WSP paper in preparation).

In multiple countries, there is moderate or greater agreement amongst the following variables: presence of soap anywhere in the home, presence of soap and water at the handwash-ing place used after defecation, and presence of soap and water at the handwashing place used before food prepara-tion, suggesting internal consistency between these various observed measures (Ram, WSP paper in preparation).

For the indicator of availability of handwashing materials within one minute of request, the time frame of one min-ute is arbitrary and does not reflect the diversity of living set-ups found around the world; i.e., soap may be at the cooking place, which is located at some distance from both the latrine and the main living area of the household. We have found conflicting associations between rapid retrieval of soap and observed handwashing behavior: in Peru, there was no association; in Senegal, caregivers of young children were more likely to wash hands if they lived in households where soap was retrieved within one minute, compared to caregivers in households where soap was retrieved more slowly. In contrast, and inexplicably, in Vietnam, caregiv-ers in households where soap was retrieved within one minute were significantly less likely to be observed wash-ing hands than caregivers in households where soap was retrieved slowly. These conflicting findings indicate that rapid retrieval of soap is not currently a reasonable marker of handwashing behavior.

Comparison with health outcomes:There is some evidence for the health benefit associated with the presence of handwashing materials at designated handwashing locations. Luby and colleagues have dem-onstrated in studies from Bangladesh that the presence of water at a handwashing place was associated with a small but statistically significant decrease in respiratory illness episodes.46

The bottom lineGiven the relative expense of quantification of microbial hand contamination currently and challenges to its validity as a measurement of overall handwashing behavior, it is not recommended that hand contamination tests be built into routine evaluations of handwashing promotion programs at this time. Further study to refine microbiology as a measure of overall handwashing behavior may enhance the utility of this approach in the future.

Proxy Measures: Rapid Observation of Handwashing MaterialsObservations of the household can be efficient means to gather clues about the household’s handwashing behavior since they can be rapidly collected, in a large number of households, and at relatively low cost. Here, handwashing materials refer to soap and water. Rapid observations pro-vide useful information on whether or not soap is present in the home, whether and where the household has a des-ignated place for handwashing, whether the tools required (i.e., soap and water, or mud/ash and water) are simultane-ously in place to practice the behavior for the individual that chooses to do so, and whether soap is readily available for handwashing (e.g., within one minute).

ValidityComparison with other handwashing measures:Analyses by Halder, Luby, and colleagues have demon-strated that observation of water at the handwashing place used after defecation was associated with observed hand cleanliness,43 and with observed handwashing with soap during structured observation.44 Soap availability at the handwashing place used after defecation was also associ-ated with observed hand cleanliness.45 Similarly, in endline surveys from the Impact Evaluation of the Global Scaling Up Handwashing project, we found that presence of soap and water, at the places used to wash hands after defecation or before food preparation, was associated with observed handwashing with soap in Peru and Vietnam; in Senegal,

43 Halder et al. 2010.44 Luby, Halder et al. 2009.45 Halder et al. 2010.46 Luby and Halder 2008; Luby et al., Using Child Health Outcomes, 2011.




Use in evaluation of handwashing programsSeveral intervention studies have demonstrated immediate increases in the maintenance of handwashing facilities,49 including studies in Bangladeshi households,50 Kenyan schools,51 and community health clubs.52

Other useful informationHand cleansing agents include soap, ash, or mud, depend-ing on the cultural context and the focus of the hand-washing promotion program (e.g., soap specifically or any cleansing agent). Choice of specific indicators should be made based on the behavioral recommendations included in the handwashing promotion.

The bottom lineGiven the associations between presence of soap and water at designated handwashing locations and observed hand-washing behavior in multiple countries, and the inverse association between presence of water and respiratory ill-ness in two studies in Bangladesh, as well as the efficiency of collecting these data, rapid observation of handwashing materials—anywhere in the home and especially at desig-nated handwashing locations—is an important approach to measuring handwashing behavior.

Proxy Measures: Hand Cleansing Product ConsumptionProduct consumption has been used as an approach to mea-sure hand hygiene in healthcare settings in high-income countries for many years.53 Studies have measured the vol-ume of soap or hand sanitizer in a fixed container at two time points in order to estimate the volume of cleansing agent consumed in the duration between measurements.

Although common in healthcare settings in high-income settings, there is little information about estimation of soap consumption as a measure of handwashing behavior in low- and middle-income countries. One approach to

EfficiencyRapid observations are being widely used to capture handwashing behavior, including in the Multi-indicator Cluster Surveys (MICS) supported by UNICEF. The MICS 4 questionnaires include the following indica-tors: designated place for handwashing where soap and water are present, and availability of soap anywhere in the dwelling. The Rapid CATCH indicators used by the US Agency for International Development (USAID) child survival grantees include the measurement of the presence of soap at the location where hands are usually washed (http://www.childsurvival.com/kpc2000/kpc2008.cfm).

Potential for bias or data collection errorsIn Bangladesh, Gadgil and colleagues made serial visits—in all 1,716 visits—to 220 households to assess the presence (and weight) of soap and other household toiletries.47 Al-though a majority of households were found to have soap at any one visit, only about 50 percent of households had soap available at every visit.

The presence of soap and water at a designated handwash-ing place cannot confirm the frequency or consistency of handwashing with soap for the individual or the house-hold as a whole, whether hands are washed during critical times such as after defecation. While the presence of soap is necessary for the behavior (handwashing with soap) to be carried out, it is not sufficient, in that a host of logisti-cal and psychosocial determinants likely induce or prevent the handwashing behavior from being carried out.48 Addi-tionally, rapid observations of the household do not provide information on the handwashing behavior of an individual of interest, such as the mother of a young child. Instead, they may only provide information about the household as a whole, since, in most households, soap is a communal resource and not an individual one. Still, since handwash-ing behavior tends to be socially mediated, household-level measurement may be very useful for describing handwash-ing behavior of a population.

47 Gadgil et al. 2010.48 Danquah 2010.49 Vindigni et al. 2011.50 Luby, Agboatwalla et al. 2009.51 Freeman et al. 2011.52 Whaley and Webster 2011.53 McGuckin et al. 2009.



www.wsp.org 9

Potential for bias or data collection errorsHuman error in the collection of consumption data, e.g., due to poor training on use of scales to weigh soap or lack of standardization in volume measurement, represents an important source of bias in these data.

Use in evaluation of handwashing programsLuby and colleagues found no difference between hand-washing intervention and control households with respect to soap purchase, despite seeing differences between com-parison groups with respect to presence of soap and water at a handwashing place, and rubbing of hands several times during a handwashing demonstration.55

Other useful informationIn Gadgil’s study of soap consumption in Bangladesh, the median bar soap weight consumption was estimated at 1.2 grams/day per household resident; laundry soap consumption was estimated at 2.3 grams/day per house-hold resident.56 Given the lack of data on soap consump-tion from households in low-income settings, Gadgil’s data may serve as a useful reference.

The bottom lineThere is insufficient evidence for or against the use of soap expenditure to make a recommendation on this measure at this time. A number of studies measuring soap consump-tion, through volume or weight checks, are currently in the analysis phase. Whether soap consumption should be used more broadly to measure handwashing behavior will be borne out by these additional analyses. Measuring con-sumption by evaluating soap weight or volume requires two visits to the home or facility, thereby increasing the costs and logistical challenges of the data collection.

Proxy Measures: Observation of Behavior During Handwashing DemonstrationOne approach to using rapid observations to obtain clues to individual behavior is to ask the individual of interest to demonstrate usual handwashing practice, in general, or

measuring soap consumption is to estimate the amount of money spent by households on soap. Alternatively, Gadgil and colleagues investigated the utility of serial soap weight measurement as a method to estimate household soap con-sumption, which was defined as the difference between soap weights in two serial visits. Soap weight differences were stable across the approximately eight visits made to each household, suggesting that one estimate of soap weight dif-ference, i.e., calculated from weights measured at two serial visits two to three days apart, is sufficient to estimate soap consumption. Wealth was associated with the presence of soap in the home but was not associated with soap weight differences.

ValidityComparison with other handwashing measures:In the study by Gadgil and colleagues, consumption of bar soap was correlated with consumption of laundry de-tergent, the total number of handwashing stations in the home, and the number of handwashing events by the main caregiver as observed during structured observation.54 To our knowledge, this is the only study to date reporting find-ings on the use of serial soap weights for estimation of soap consumption at the household level in a low- or middle-income country.

Comparison with health outcomes:At the time of this writing, we are not aware of studies that have assessed the relationship between soap consumption and health outcomes in low- or middle-income country contexts.

EfficiencyThe minimum of two visits required to estimate change in soap weight or volume reduces the efficiency of data collec-tion and, thus, this approach may not be feasible for many large-scale studies, such as DHS or MICS.

Amount spent recently on soap purchase can be easily que-ried in a questionnaire.

54 Gadgil et al. 2010.55 Luby, Agboatwalla et al. 2009.56 Gadgil et al. 2010.




Sagerman and colleagues found that the duration of hand rubbing increased from baseline to post-intervention even among study participants that were not instructed to in-crease the duration of hand rubbing, suggesting reactivity to the observation.60 A similar finding was noted among schoolchildren in a study of waterless hand cleansing in Nairobi, Kenya.61

Use in evaluation of handwashing programsIn a study evaluating residual effects of a handwashing promotion intervention 18 months after intervention, Luby and colleagues showed that subjects in intervention households were more likely to rub palms several times during a handwashing demonstration, compared to sub-jects in control households.62 Freeman and colleagues reported that Kenyan students, exposed to a water treat-ment and hygiene promotion intervention, completed the several steps of handwashing during a handwash-ing demonstration, compared to students in control schools.63

The bottom line To date, relatively few studies have examined the validity of handwashing demonstrations for the measurement of usual handwashing behavior. There is some evidence that these data are compromised by social desirability bias but fur-ther exploration is required to understand the utility of this measurement approach.

Proxy Measures: Visual Inspection of Hand CleanlinessA number of studies have used visual inspection of respon-dent hands to characterize their degree of cleanliness. Typi-cally, a three-point scale has been used, denoting “clean,” “no visible dirt but unclean appearance,” and “visible dirt.”

In some settings, subjects have expressed concern about whether hand inspections are appropriate and acceptable. The evaluator should decide whether hand inspections are appropriate in the setting of the study or evaluation. The

as typical after defecation. Details to note include washing of one or both hands, use of soap or other hand cleansing agent, duration of lathering or hand rubbing, and method of drying (or not drying).

ValidityComparison with other handwashing measures:In the 2008 study by Biran and colleagues, there was a fair degree of agreement between observation of soap use dur-ing handwashing demonstration and observation of both hands being washed with soap after all fecal contact events witnessed during structured observation.57

As with a number of measures of handwashing behavior, awareness of social desirability may prompt improved handwashing practice during the demonstration compared to usual behavior. Halder and colleagues found that, while 51 percent of caregivers used soap when asked to demon-strate how they typically wash their hands after defecation, only 33 percent of caregivers were observed to wash hands during structured observation.58

Comparison with health outcomes:Use of soap during demonstration of usual handwashing after defecation was significantly associated with less di-arrhea than non-use of soap in one study.59 Air drying of hands during the handwashing demonstration was also in-dependently associated with reduced respiratory illness in this same study.

EfficiencyObservation of behavior during handwashing demonstra-tions is typically quite efficient since it can be incorporated into an interview with the target respondent. A few minutes may be required for the respondent to gather any necessary materials to perform the handwashing.

Potential for bias or data collection errorsIn a randomized controlled trial investigating responsiveness to simple versus more complex handwashing instructions,

57 Biran et al. 2008. 58 Halder et al. 2010.59 Luby et al., The Effect of Handwashing, 2011.60 Sagerman et al. 2011.61 Pickering, Davis et al. 2011.62 Luby, Agboatwalla et al. 2009.63 Freeman et al. 2011.



www.wsp.org 11

caregivers (Ram, baseline report). In all three countries, households of caregivers with clean hands were more likely to have had soap and water present at a place designated for handwashing before food preparation and eating. Also, in Senegal and Vietnam, households of caregivers with clean hands were significantly more likely to have had soap and water observed at the place for handwashing after defeca-tion. Caregivers with clean hands were progressively and significantly more likely to be in the wealthiest four quin-tiles than in the poorest quintile in all three countries.

Halder and colleagues found that household wealth and water availability at handwashing locations were signifi-cantly associated with the cleanliness of mother’s and child’s hands.66 The strong associations between wealth and hand cleanliness, seen in Halder’s work and ours in the Global Scaling Up Impact Evaluation, may be attributable to un-measured determinants of handwashing behavior or to interviewer bias, should be considered in analysis of these data generated from observational studies.

Other useful informationIn some settings, subjects have expressed concern about whether hand inspections are appropriate and acceptable. The evaluator should decide whether hand inspections are appropriate in the setting of the study or evaluation. The use of photos, pictorials, or direct observation of hands repre-senting each level of cleanliness may be helpful to improve on standardization of data collection. The trainer should seek to ensure high inter-rater reliability; i.e., different enumerators code the same level of cleanliness for a pair of hands.

The bottom line Visual inspections of hand cleanliness are efficiently per-formed. This measure is relatively novel and needs to further exploration in order to determine whether it is valid globally.

Direct Measures: Structured ObservationsStructured observations consist of the placement of an ob-server for several hours, typically between three and seven hours, in a household.67 The observer records opportunities

use of photos, pictorials, or direct observation of hands repre-senting each level of cleanliness may be helpful to improve on standardization of data collection. The trainer should seek to ensure high inter-rater reliability; i.e., different enumerators code the same level of cleanliness for a pair of hands.

ValidityComparison with other handwashing measures:Pickering and colleagues have found that visible dirt on palms, fingerpads, or under nails is associated with in-creased microbiological contamination of hands.64

A high hand cleanliness score was significantly associated with observed handwashing with soap after fecal contact in both Peru and Vietnam, after adjustment for wealth, indicating that hand cleanliness inspection is a reasonable proxy measure for handwashing behavior in some contexts; we did not find such an association in Senegal (Ram, WSP paper in preparation).

Comparison with health outcomes:Luby and colleagues have found that observation of visibly clean finger pads on a child’s hands was associated with reduced diarrhea prevalence; there was no association be-tween mother’s hand cleanliness and child’s diarrhea risk.65

EfficiencyVisual inspections of hand cleanliness are efficiently per-formed, easily included in interviews with the target respon-dent. Standardization of fieldworkers’ coding of cleanliness is critical, and should be done using photographs of local individuals’ hands. Data for the measure may be collected surreptitiously, or by actively informing the respondent of the examination. If the latter approach is chosen, it is pos-sible that the participant will refuse the inspection.

Potential for bias or data collection errors In the Impact Evaluation of the Global Scaling Up Hand-washing project, at baseline, clean hands were identified in 42 percent of 3,718 caregivers in Peru, 71 percent of 1,992 Senegal caregivers, and 63 percent of 3,068 Vietnam

64 Pickering, Davis et al. 2010.65 Luby et al., Using Child Health Outcomes, 2011.66 Halder et al. 2010.67 Bentley et al. 1994; Curtis et al. 2001; Biran et al. 2008.




their hands, children living in households where the food preparer washed at least one hand with water only (odds ratio [OR] � 0.78; 95% confidence interval [CI] � 0.57 � 1.05), washed both hands with water only (OR � 0.67; 95% CI � 0.51 � 0.89), or washed at least one hand with soap (OR � 0.30; 95% CI � 0.19 � 0.47) had less diar-rhea.”73 To our knowledge, this is the first evidence from household settings that handwashing at two specific criti-cal times is associated with disease risk. There was no asso-ciation between observed handwashing behavior at other critical times (before feeding a child, before eating, after cleaning the anus of a child who had defecated) and diar-rhea risk. This study highlights the importance of captur-ing detailed information around handwashing behavior and underscores the utility of structured observation for such crucial data collection.

EfficiencySeveral groups have successfully completed structured ob-servations on the scale of hundreds of households with-out substantial difficulty. They provide a wealth of detail regarding handwashing behavior at critical times of inter-est, including defecation, feeding, eating, and cooking. Be-cause of the rich contextual detail, structured observation data can be used to explore thoroughly the determinants of handwashing behavior. Aunger and colleagues used struc-tured observation data to elucidate psychosocial determi-nants of handwashing behavior, and found the following to be observed with improved handwashing behavior: habitual handwashing at critical times, lack of concern about the costliness of soap, and an expressed need for cleanliness of oneself or one’s household.74

The use of structured observation for measurement of handwashing behavior can incur substantial costs in terms of personnel time. Highly trained staff experienced in be-havioral observation methods are preferable for performing observations, or at least training observers.75

for handwashing, such as feeding a child or visiting the toilet, and the target respondent’s handwashing practices. A sample tool used in the healthcare setting is provided by McAteer and colleagues;68 sample tools used in school and community settings are available from the author of this paper. The benefits of structured observation are the ability to record objective data on handwashing practices and the richness of information gathered.69 During struc-tured observation, the observer has the opportunity to re-cord information about numerous individuals of interest, including mothers, young children, non-caregiver males, etc. Additionally, the observer can record detailed informa-tion on particular critical times, whether hands are washed, whether both hands are washed, the type of cleansing agent used, and the way in which hands are dried. This richness of details allows for assessment of consistency in handwashing practices. Respondents may be assigned to categories repre-senting degrees of appropriate handwashing practice, based on observation of behavior during multiple opportunities for handwashing.70

ValidityComparison with other handwashing measures:Structured observation data has frequently been used as the standard of comparison for other handwashing measures in a number of studies, as noted in other parts of this document.71

Comparison with health outcomes:A recent robust analysis of structured observation data supports the utility of structured observation for detection of handwashing measures that are meaningful for disease risk. Luby and colleagues examined diarrhea prevalence among children of caregivers in 347 households observed during five-hour structured observations.72 This study in-dicated that observation of handwashing with soap after defecation was significantly associated with lower likeli-hood of diarrhea. “Compared with children living in households where persons prepared food without washing

68 McAteer et al. 2008.69 Bentley et al. 1994.70 Biran et al. 2008.71 Curtis et al. 2001; Biran et al. 2008; Biran et al. 2009; Luby, Halder et al. 2009; Halder et al. 2010.72 Luby et al., The Effect of Handwashing, 2011.73 Ibid.74 Aunger et al. 2010.75 Bentley et al. 1994.



www.wsp.org 13

Despite the potential for reactivity, structured observa-tions typically demonstrate very poor handwashing be-havior among study populations. Curtis and colleagues have shown that the percent of toilet use events followed by handwashing with soap ranged from 3 percent to 42 percent, with a mean of 17 percent for the 11 low- and middle-income countries under study.78 Handwashing with soap was performed on an average of 5 percent of occasions of feeding an index child. Similarly, in the three Global Scaling Up Handwashing project countries, we find that fewer than 10 percent of food preparation events and child feeding events are preceded by handwashing with soap. Despite the unarguable presence of reactivity to structured observation, structured observations tend to confirm the tremendous need to improve handwashing behaviors in low- and middle-income countries.

Use in evaluation of handwashing programsStructured observations have been used for the evaluation of handwashing promotion programs. The risk with reac-tivity to structured observation is that the evaluator would overestimate the change in handwashing behavior in re-sponse to an intervention. Biran and colleagues evaluated a hygiene education campaign in rural India and found no improvement in observed handwashing after defecation, cleaning a child who had defecated, or other fecal contact, despite about one-third of respondents having heard of the intervention.79 Huda and colleagues found similarly that a large-scale program to promote water, sanitation, and hy-giene behaviors did not result in improved handwashing behaviors, as evidenced by structured observations.80

Other useful information It is essential that the researcher/evaluator intensively train observers to record in a standardized fashion. Such train-ing should, for example, emphasize details such as the time frame within which handwashing would be considered as-sociated with a particular critical time (e.g., the number of minutes after defecation that handwashing occurs) as well as observational techniques, such as the need to utilize neutral body language and avoid judgment or prompting of socially

In a 2007 study in Bangladesh commissioned by the Water and Sanitation Program, where five-hour struc-tured observations were compared to 90-minute structured observations, shorter observation periods resulted in a greater-than-proportional loss in observed numbers of fecal contact events, suggesting that it is counter-productive to shorten the observation periods in an attempt to reduce data collection costs. Extended durations of observation come at a cost since each observer can only conduct one observation per household per day. It would be practically difficult to carry out two five-hour, or even four-hour, ob-servations in a single day without risking substantial inter-viewer fatigue and compromise in the quality of observed data. Moreover, handwashing, bathing, and toileting be-havior may differ according to the time of day, possibly rendering morning observations incomparable to afternoon observations. Thus, depending on the required sample size to demonstrate project outcomes of interest, struc-tured observations would require substantial numbers of trained individuals, or a prolonged data collection period, either of which might be expensive for the program being monitored.

Potential for bias or data collection errors The same awareness of social desirability that likely results in overestimation of self-reported handwashing behavior may also result in reactivity to the presence of the observer during a structured observation.76 That is, an individual may practice better handwashing behaviors while an ob-server is present than when she is unobserved. Using accel-eration sensors in soap bars, we have shown that households increased the number of times soap was used by 35 percent during structured observation, compared to days when no human observer was present.77 Nearly one-quarter of households, characterized by markers of improved socio-economic status, increased soap use by more than double during structured observation. Reactive households were also much more likely to have soap available at a desig-nated handwashing location near the toilet, suggesting that they may, indeed, prioritize hand hygiene more than non-reactive households.

76 Cousens et al. 1996.77 Ram, Halder et al. 2010.78 Curtis et al. 2009.79 Biran et al. 2009.80 Huda et al. 2011.




are time limitations of having a human observer present (e.g., five hours or seven hours before an observer fatigues). While, theoretically, a human observer may miss events if many are occurring simultaneously, video can be viewed repeatedly in order to capture the details of most or all events recorded. Still, videos for quantitative evaluation have typi-cally been recorded by placing cameras in fixed locations; as key targets of observation (e.g., primary caregivers) move about from place to place during their daily activity, video cameras in fixed locations would likely miss a number of events that could be captured by the human observer who would be able to move about to gain appropriate views of the target’s handwashing behavior.

ValidityComparison with other handwashing measures: To our knowledge, there are no published comparisons of video-based handwashing measurement to other approaches to measuring the behavior in low- and middle-income coun-try settings. In a school-based study in Kenya, Pickering and colleagues have found the rate of hand cleansing with waterless sanitizer to be roughly similar among schoolchil-dren provided access to alcohol-based hand sanitizer (hands cleansed with sanitizer during 82 percent of events detected by structured observation, 79 percent of events detected by video camera).81 On preliminary data analysis, video sur-veillance captured approximately 30 percent fewer events than structured observation, perhaps since the camera was focused on a fixed location (sanitizer dispenser), rather than being responsive to movement of individuals.

Comparison with health outcomes:To our knowledge, there is no data from community set-tings in low- and middle-income countries on the relation-ship between video-recorded handwashing behavior and health outcomes.

EfficiencyVideo may best be used to document handwashing behav-ior in particularly densely populated areas, e.g., outside shared latrines in a community, in school or healthcare set-tings, since a large number of events could be captured in such contexts. Researchers should be aware of the intensity

desirable behaviors). Ideally, the time frame for observation would be based on local knowledge of the timing of behav-iors of interest. If handwashing after defecation is the behav-ior of interest, then timing the observation to ensure that the observer is present in the home when most people defecate (early in the morning in many cultures) would be very im-portant. This is not always feasible due to safety or logistical concerns, thus necessitating structured observation at other times of day. Over the span of several hours, an observer can only complete observation in one household.

A key factor in minimizing reactivity to structured obser-vation is the information given to the target respondent in advance of the observation. During the informed consent process, and in other verbal and non-verbal communications by the study worker and observer, it is critical not to em-phasize that the observation is principally about measuring handwashing behavior. Without violating the respondent’s right to understand the nature of the study, the evaluator may indicate that the observation is aimed at understanding “general household practices.” The observer’s training and skills with respect to objective data collection may also im-pact reactivity on the part of the individuals being observed.

The bottom line Information obtained from structured observation can guide handwashing promotion recommendations but also inform an evaluator about whether a handwashing pro-gram is affecting behaviors of interest, and that are rel-evant to child health goals. Structured observations have been used to evaluate the behavioral impact of hand hy-giene promotion programs and indicate that, while there may be reactivity to structured observation, that reactivity may not be enhanced solely by exposure to handwashing promotion.

Direct Measures: Video ObservationVideo observation has been used in several studies pub-lished or carried out since 2008, in quantitative studies in schools and health facilities and qualitative studies in households. A large amount of data can be captured by video for two reasons: capturing of events is not restricted as structured observation might be, since in the latter, there

81 Pickering, Davis et al. 2011.



www.wsp.org 15

The ethical considerations are considerable with the use of video recording. Under traditional human subjects’ research guidelines, a single human observer is bound to the pri-vacy of individuals that she is observing. Video recordings that can be preserved indefinitely in digital media can pose privacy and confidentiality concerns to those under obser-vation. A subject’s identity can be disclosed by the video. Individuals may display private, or even illegal, behavior whether they are aware of the video cameras in place or not. Researchers are reminded to solicit scientific and ethi-cal reviews from accredited agencies carrying out human subjects research oversight before attempting to use obser-vation methods. It is ethically appropriate, and relevant to the program’s best interests, to maintain the confidentiality of video data, as with all other identifiable data. Procedures regarding the handling of video displaying potentially pri-vate or illegal behavior should be defined a priori.

The bottom line Video recording remains a relatively novel approach to measuring handwashing behavior. It may be most useful for evaluating behavior in semi-private or public settings, such as in schools and healthcare facilities, where there is minimal expectation of privacy. The evaluator seeking to use video observation needs to plan not only for captur-ing the video data, but also the extensive time required to analyze such data.

Direct Measures: SensorsThis section describes sensor-based methods of recording handwashing behavior. One such method is a Unilever-developed technology embedding accelerometers in ordi-nary-appearing Lifebuoy® soap. The accelerometer tracks movement of the soap in three dimensions.83 Based on the movement patterns of the soap, the number of times soap is used in a given time period can be counted. The soap can be left in a household for several days, allowing for ob-servation of soap use behavior over a much longer period of time than would be feasible by structured observation. The accelerometer’s ability to detect consistency in soap use behavior has been demonstrated in Bangladesh, where the

of person-time required to analyze video observation data. As in the Armellino study, we found that a large number of events are captured in video in schools because of the large number of persons under observation (Pickering, unpub-lished observations). Thus, even by reviewing video at 8� or 16� speed, many person-months were spent analyzing data from just a few weeks of observation.

Potential for bias or data collection errors We suspect, based on our work using video cameras in schools in Kenya, that there is reactivity to the presence of video cameras—e.g., children looking at the cameras and showing amusing behavior (Pickering, unpublished observations).

Use in evaluation of handwashing programsArmellino and colleagues placed video cameras positioned to capture hand cleansing behavior at every sink and sani-tizer dispenser in 11 patient rooms in an intensive care unit in New York State, USA.82 In a 16-week period before any intervention, 60,542 events were captured and rates of hand cleansing were reported to be less than 10 percent. After feedback of video-recorded data was provided to the health workers, 73,080 events were observed and hand cleansing was estimated at 82 percent based on video obser-vation. This study indicates that video observation of hand cleansing behavior during patient contact was responsive to an intervention that provided feedback about individuals’ hand hygiene behavior.

Other useful informationVideo cameras may be particularly useful to establish up-take of handwashing hardware and purposively mounted in fixed locations through an intervention program. In house-holds, schools, or health facilities that have not received such hardware, or that have not received guidance on place-ment of such hardware, video recording may be less mean-ingful; the location of the hardware relative to the latrine or food preparation area may be distant, or inconvenient, making the link between the critical time and the associated handwashing behavior difficult to establish.

82 Armellino et al. 2012. 83 Ram, Halder et al. 2010.




the reach of most researchers and handwashing program evaluators.

Even more distant is the potential for computer-aided tracking of humans and objects, which may be useful as a detection methodology or even a technology to aid in re-minding individuals to wash hands if they have not done so at critical times.87 We see little potential in the near future for such interesting technologies in the low- and middle-income country settings of concern here.

ValidityComparison with other handwashing measures: There is minimal information on the comparison of the sensor-based methods to other handwashing measures in low- and middle-income settings. An early study of the accelerometer-embedded soaps demonstrated a significant correlation between soap consumption and number of soap uses, as indicated by the accelerometers.88

Comparison with health outcomes:We are not aware of health outcomes data from resource-limited settings for sensor-based measures of handwashing behavior.

EfficiencyCurrently, the largest drawback to any of the sensor-based ap-proaches described to date is the limitation of capacity to deploy, extract, and analyze data to highly qualified research staff, often to those involved in developing the sensor methodology. Trans-fer of technical capacity, or development of more field-friendly sensor approaches, are imperative if any of the sensor-based methods are to become more widely adopted for monitoring handwashing behavior. However, such transfer clearly comes at the risk of lack of assurance of the various steps contributing to data quality. The cost of individual sensors is expected to be high; for example, the accelerometers for embedding into soap are projected at approximately US$120, which may be prohib-itive if a program actually had to purchase large numbers of the sensors.

number of times soap was used in a household was remark-ably consistent across each of eight days. The accelerometer allows detection of soap movement, with a time stamp, but by itself, it does not allow for describing handwashing at specific critical times. Biran reports attaching motion sen-sors to water vessels reserved for cleansing oneself after def-ecation, thereby allowing for detection of defecation events and subsequent handwashing.84 This novel approach found that, despite an increase in the number of soap uses overall, there was no increase in the number of soap uses following a defecation event among intervention households, com-pared to control households.

As an alternative sensor-based approach, colleagues at the London School of Hygiene and Tropical Medicine have used wireless infrared sensors at the entryways of restrooms, and at soap dispensers inside restrooms, to measure soap use among British commuters.85 The sensors at the entryways of restrooms allowed for detection of entry into the rest-room, such that the denominator was the number of people entering the restroom in a given hour. The measurement of handwashing behavior was based on the numerator defined as the number of soap dispenses in a given hour, with uses in the first five minutes of the hour excluded. This novel approach enabled the investigators to investigate the imme-diate effects on soap use behavior of automated text messages appearing at the entryway of the restrooms that reinforced a number of psychosocial motivators of handwashing. The cost and sophisticated analyses required would place this technology, as with soap with motion sensors, squarely in the research arena for the foreseeable future.

Another technology-based approach is the use of radio-frequency-controlled transmitters and readers that can demonstrate proximity between an individual wearing a transmitter and a hand cleansing device bearing a reader; such an approach can yield individual-level data on the use of handwashing devices.86 We are not aware of the use of such devices in low- and middle-income settings and antici-pate that cost and analysis considerations put them beyond

84 Biran et al. 2009.85 Judah et al. 2009.86 Munro and Munro 2009.87 Hoey et al. 2010.88 Ram, Halder et al. 2010.



www.wsp.org 17

and analysis, are required before sensor-based technologies can be readily adopted for handwashing measurement in resource-limited settings.

Use of Composite Measures Several studies have attempted to combine information from different methods of measuring handwashing behav-ior.91 These composite measures have included information obtained from demonstration of handwashing, as well as information from self-report. Strina and colleagues used a composite measure that included handwashing informa-tion, as well as observations of other hygiene behaviors, such as household drinking water storage and treatment, washing of vegetables, eating potentially dirty food, and so on.92 In our review of the literature, no composite measure of handwashing has been validated in different geographic or cultural contexts.

ValidityComparison with other handwashing measures:Stevenson and colleagues have performed a series of studies in Australia to evaluate an index of responses to 23 ques-tions (HI23) measuring self-reported hygiene behaviors.93 The questions address several domains of hygiene, includ-ing hand hygiene, food hygiene, and home hygiene. Steven-son shows that the scale is associated with a disgust scale, as well as observed handwashing behavior in response to disgust-eliciting stimuli. We are not aware of replications of these studies, or adaptation of Stevenson’s hygiene index to low- and middle-income country settings, where hygiene-related attitudes and perceptions of social desirability may differ vastly from the Australian setting of Stevenson’s work.

Gadgil and colleagues tested a 14-point scale for handwash-ing habit in a study of soap consumption and use in urban Bangladesh, based on the Verplanken self-reported habit index (evaluated by Verplanken for various habits other than handwashing).94 They found no association between handwashing habit and consistent presence of soap in the home.

Potential for bias or data collection errors It is possible that, overall, households increase the number of times they use soap when the sensor is in place, compared to when the sensor is not in place. We did not find evidence that the households used the accelerometer-embedded soap more frequently on the first day of its presence in the home and then reduced use as they became accustomed to it.89

Use in evaluation of handwashing programsBiran and colleagues have used the accelerometer-embed-ded soap to evaluate a handwashing promotion program in rural India and found an increase in the number of soap uses overall.90

Other useful informationSensor-based data may provide individual-level informa-tion, as with radiofrequency tags, which would be worn by individuals of interest, or group-level information, as with the accelerometer, which is embedded in a bar of soap that may be used by the entire family.

In many countries, households often use multipurpose bar soaps, liquid soaps, or powdered detergent for washing hands. Therefore, embedding accelerometers multipurpose bar soaps, which may also be used for washing laundry or dishes, is potentially problematic in terms of identifying handwashing-specific events. The movement signatures for handwashing may be difficult to distinguish from those of washing clothes, washing dishes, or playing with the soap. Replacing powder or liquid with a bar may allow introduce substantial bias, in that the observed household is given a novel and “special” way of washing hands that it has not previously used.

The bottom line These various sensor technologies are in the early stages of development and, to our knowledge, none are readily deployable for routine research or evaluation purposes in low- or middle-income countries. Further validation, and a substantial lowering of costs and expertise required for use

89 Ram, Halder et al. 2010.90 Biran et al. 2009.91 Hoque et al. 1995; Yalcin et al. 2004; Sandora et al. 2005.92 Strina et al. 2003.93 Stevenson et al. 2009.94 Verplanken and Orbell 2003; Gadgil et al. 2011.




Other useful informationThe interpretation of an index is often difficult and com-parison of index scores across studies or evaluations may be challenging or impossible. For example, an index con-structed using principal component analysis, a data reduc-tion method, would be meaningless outside of the study sample within which it is developed.

The bottom lineAt present, there is insufficient evidence to support the use of a single composite measure in research or evaluation of handwashing promotion. This is an area that merits explo-ration in secondary analyses of large datasets containing numerous measures of reported and observed measures of handwashing behavior.

Comparison with health outcomes:Webb and colleagues constructed individual and combined hygiene indices using observed information on drinking water, food, personal hygiene, and household domestic hy-giene. Scores for the overall hygiene index and the personal hygiene index (which included wearing of shoes, and ob-servations of cleanliness of mother and child hands) were inversely associated with child diarrhea morbidity.95

Potential for bias or data collection errorsStevenson’s hygiene index score was correlated with scores indicating the social desirability of hygiene, but awareness of social desirability explained only 2.5 percent of the vari-ance in the hygiene index score, indicating relatively small degree of bias.96

95 Webb et al. 2006.96 Stevenson et al. 2009.


www.wsp.org 19

This document attempts to give a balanced view of each of the routinely used and novel methods of measuring hand-washing behavior. Positive and negative attributes of each method have been described (see Table 1). The attempt to be balanced may lead to skepticism about the utility of measur-ing handwashing behavior at all. Since there are few perfect measurements available for outcomes of human behavior or health, researchers and public health practitioners must fre-quently accept the limitations of the measures available to them, but not get paralyzed by those limitations. Examples of imperfect measures that still provide useful and necessary information are self-reported breastfeeding or self-reported use of oral rehydration therapy for diarrhea treatment, as well as caregiver reported symptoms of childhood diar-rhea and other illnesses. These measures are undertaken in

every Demographic and Health Survey (DHS) and every Multiple Indicator Cluster Survey (MICS) undertaken in resource-poor countries. While these self-reported measures may exaggerate individuals’ true practices and health condi-tions, they do provide insights into trends in these behav-iors over time and important predictors of child morbidity and mortality. To that end, described below are potential approaches to measuring handwashing behavior for a vari-ety of program types and settings. Reference to health out-comes as proxy measures of handwashing behavior has been intentionally minimized, since, almost universally, mea-surement of health outcomes such as diarrhea incidence or prevalence is very costly. As indicated below, additional data on the relationship between other measures of handwash-ing behavior and health outcomes is clearly needed.

ConclusionIII.


Practical Guidance for Measuring Handwashing Behavior: 2013 Update Conclusion


TAB

LE

1:

SU

MM

AR

Y O

F S

TR

ATE

GIE

S T

O M

EA

SU

RE

HA

ND

WA

SH

ING

BE

HA

VIO

R

Mea

sure

men

t S

trat

egy

Req

uire

men

ts f

or

Use

Ad

vant

ages

Dis

adva

ntag

esA

pp

rop

riate

Set

ting

for

Use

Sel

f-re

po

rt•

Que

stio

nnai

re•

Trai

ning

• Is

effi

cien

t•

Sho

wn

to b

e as

soci

ated

w

ith h

ealth

out

com

es•

Can

be

inco

rpor

ated

into

m

ultip

urp

ose

surv

eys

• S

how

n re

pea

ted

ly t

o ov

er-

estim

ate

hand

was

hing

b

ehav

ior

• Q

uest

ionn

aire

s m

ay b

e us

eful

for

trac

king

kn

owle

dge

, mon

itorin

g th

e re

ach

of a

han

d-

was

hing

pro

mot

ion

pro

gram

, or

iden

tifyi

ng

bar

riers

to

hand

was

hing

Mic

rob

io-

log

ical

Han

d

Co

ntam

inat

ion

• S

upp

lies

for

hand

rin

se/

hand

imp

rint

colle

ctio

n•

Mic

rob

iolo

gy la

bor

ator

y or

eq

uip

men

t fo

r fie

ld-

leve

l qua

ntita

tion

of h

and

co

ntam

inat

ion

• Tr

aini

ng a

nd q

ualit

y co

ntro

l•

At

leas

t on

e st

aff m

em-

ber

with

mic

rob

iolo

gy

exp

erie

nce

• Is

ob

ject

ive

• R

efle

cts

ind

ivid

ual h

and

co

ntam

inat

ion

• Is

not

rel

iab

le•

Is e

xpen

sive

• S

erve

s as

a p

roxy

mea

sure

of

han

dw

ashi

ng b

ehav

ior

• R

equi

res

som

e st

ruct

ured

ob

serv

atio

n fo

r m

easu

re-

men

t at

crit

ical

tim

es (e

.g.,

afte

r d

efec

atio

n)

• W

ell-

fund

ed h

and

was

hing

pro

mot

ion

pro

-gr

am e

valu

atio

ns o

r re

sear

ch s

tud

ies

may

b

e ab

le t

o im

pro

ve t

he u

tility

of m

icro

bio

-lo

gica

l han

d c

onta

min

atio

n as

a m

arke

r of

ha

ndw

ashi

ng b

ehav

ior

• N

ot r

ecom

men

ded

for

smal

l or

min

imal

ly

fund

ed h

and

was

hing

pro

mot

ion

pro

gram

s•

Not

rec

omm

end

ed fo

r na

tiona

lly o

r re

gion

-al

ly r

epre

sent

ativ

e m

ultip

urp

ose

surv

eys

(e.g

., D

HS

, MIC

S)

Rap

id

Ob

serv

atio

ns o

f H

and

was

hing

M

ater

ials

• C

heck

list

• Tr

aini

ng•

Is e

ffici

ent

• Is

ob

ject

ive

• C

an b

e in

corp

orat

ed

into

mul

tipur

pos

e su

rvey

s•

Sho

wn

to h

ave

inte

rnal

va

lidity

with

str

uctu

red

ob

serv

atio

n

• S

erve

s as

a p

roxy

mea

sure

• C

anno

t co

nfirm

the

fre-

que

ncy

or c

onsi

sten

cy o

f ha

ndw

ashi

ng•

Doe

s no

t re

flect

ind

ivid

ual-

leve

l beh

avio

r

• W

ell-

fund

ed h

and

was

hing

pro

mot

ion

pro

-gr

am e

valu

atio

ns o

r re

sear

ch s

tud

ies

• E

valu

atio

ns o

f sm

all o

r m

inim

ally

fund

ed

hand

was

hing

pro

mot

ion

pro

gram

s•

Nat

iona

lly o

r re

gion

ally

rep

rese

ntat

ive

mul

tipur

pos

e su

rvey

s (e

.g.,

DH

S, M

ICS

)

Han

d C

lean

sing

P

rod

uct

Co

nsum

ptio

n

• S

cale

to

wei

gh s

oap

, or

cont

aine

rs t

o m

easu

re

volu

me

• A

min

imum

of t

wo

visi

ts

to h

ouse

hold

s (to

mea

-su

re c

hang

e in

soa

p

wei

ght

or v

olum

e b

etw

een

visi

ts)

• C

heck

list

• Tr

aini

ng

• Is

ob

ject

ive

• R

equi

res

two

visi

ts t

o th

e ho

me,

the

reb

y re

duc

ing

effic

ienc

y of

d

ata

colle

ctio

n

• W

ell-

fund

ed h

and

was

hing

pro

mot

ion

pro

gram

eva

luat

ions

or

rese

arch

stu

die

s

Ob

serv

atio

n o

f B

ehav

ior

Dur

ing

H

and

was

hing

D

emo

nstr

atio

n

• C

heck

list

• Tr

aini

ng•

Is e

ffici

ent

• D

oes

not

rel

y on

res

pon

-d

ent

self-

rep

ort

• C

an b

e in

corp

orat

ed in

to

mul

tipur

pos

e su

rvey

s•

Sho

wn

to h

ave

inte

rnal

va

lidity

with

str

uctu

red

ob

serv

atio

n•

Ass

ocia

ted

with

dia

rrhe

a an

d r

esp

irato

ry il

lnes

s

• M

ay b

e su

bje

ct t

o re

activ

ity•

Wel

l-fu

nded

han

dw

ashi

ng p

rom

otio

n p

ro-

gram

eva

luat

ions

or

rese

arch

stu

die

s•

Eva

luat

ions

of s

mal

l or

min

imal

ly fu

nded

ha

ndw

ashi

ng p

rom

otio

n p

rogr

ams



www.wsp.org 21

TAB

LE

1:

SU

MM

AR

Y O

F S

TR

ATE

GIE

S T

O M

EA

SU

RE

HA

ND

WA

SH

ING

BE

HA

VIO

R (C

ontin

ued

)

Mea

sure

men

t S

trat

egy

Req

uire

men

ts f

or

Use

Ad

vant

ages

Dis

adva

ntag

esA

pp

rop

riate

Set

ting

for

Use

Vis

ual

Insp

ectio

n o

f H

and

C

lean

lines

s

• C

heck

list

• Tr

aini

ng•

Is e

ffici

ent

• D

oes

not

rely

on

resp

on-

den

t se

lf-re

por

t•

Can

be

inco

rpor

ated

into

m

ultip

urp

ose

surv

eys

• S

how

n to

hav

e in

tern

al

valid

ity w

ith s

truc

ture

d

obse

rvat

ion

• A

ssoc

iate

d w

ith m

icro

bio

-lo

gica

l con

tam

inat

ion

and

d

iarr

hea

pre

vale

nce

• M

ay n

ot b

e co

mp

arab

le

acro

ss s

tud

ies

and

geo

-gr

aphi

c re

gion

s•

May

be

sub

ject

ive

(bas

ed

on in

terv

iew

er t

rain

ing

and

b

iase

s)

• W

ell-

fund

ed h

and

was

hing

pro

mot

ion

pro

-gr

am e

valu

atio

ns o

r re

sear

ch s

tud

ies

• E

valu

atio

ns o

f sm

all o

r m

inim

ally

fund

ed

hand

was

hing

pro

mot

ion

pro

gram

s

Str

uctu

red

O

bse

rvat

ions

• S

truc

ture

d fo

rmat

to

cap

-tu

re d

etai

ls r

egar

din

g cr

iti-

cal t

imes

of i

nter

est

and

ha

ndw

ashi

ng b

ehav

iors

• S

ever

al h

ours

of o

bse

r-va

tion,

pre

fera

bly

at

the

sam

e tim

e of

day

in a

ll ho

useh

old

s, a

nd a

t tim

es

that

cap

ture

crit

ical

tim

es

of in

tere

st a

nd t

hat

are

lo-

cally

acc

epta

ble

• Tr

aini

ng a

nd q

ualit

y co

ntro

l•

Pre

fera

ble

to

have

sta

ff w

ith e

xper

ienc

e in

beh

av-

iora

l ob

serv

atio

n

• Is

ob

ject

ive

• R

efle

cts

ind

ivid

ual

beh

avio

r•

Cap

ture

s ric

h d

etai

l on

hand

was

hing

beh

avio

r

• S

how

n to

res

ult

in r

eact

ivity

, b

ecau

se o

f the

pre

senc

e of

th

e hu

man

ob

serv

er•

Is c

ostly

in t

erm

s of

p

erso

n-tim

e•

Req

uire

s hi

ghly

tra

ined

sta

ff

• W

ell-

fund

ed h

and

was

hing

pro

mot

ion

pro

-gr

am e

valu

atio

ns o

r re

sear

ch s

tud

ies

• S

amp

le s

ize

calc

ulat

ions

, and

con

sulta

tion

with

per

sons

with

ep

idem

iolo

gica

l and

sta

-tis

tical

exp

ertis

e, m

ay in

dic

ate

feas

ibili

ty o

f th

is a

pp

roac

h fo

r ev

alua

tion

of e

ven

smal

l ha

ndw

ashi

ng p

rom

otio

n p

rogr

ams

Vid

eo

Ob

serv

atio

ns

Vid

eo r

ecor

d-

ing

freq

uenc

y of

han

dw

ashi

ng

with

soa

p a

t cr

it-ic

al t

imes

, e.g

., af

ter

def

ecat

ion

• V

ideo

cam

eras

and

mem

-or

y ca

rds

(e.g

., S

D c

ard

s)•

Mon

itors

to

view

vid

eo

reco

rdin

g•

Use

of s

truc

ture

d fo

rmat

to

cap

ture

det

ails

reg

ard

-in

g cr

itica

l tim

es o

f int

er-

est

and

han

dw

ashi

ng

beh

avio

rs•

Sta

ff tim

e fo

r ob

serv

atio

n

• Is

ob

ject

ive

• Is

effi

cien

t•

Cap

ture

s ric

h d

etai

l on

hand

was

hing

beh

avio

r•

Vid

eo c

an b

e vi

ewed

re

pea

ted

ly in

ord

er t

o ca

ptu

re d

etai

ls m

isse

d o

n ea

rlier

vie

win

gs

• C

an r

esul

t in

rea

ctiv

ity, b

e-ca

use

of t

he p

rese

nce

of t

he

vid

eo c

amer

a•

Can

be

cost

ly b

oth

in t

erm

s of

tec

hnol

ogy

and

in t

erm

s of

per

son-

time

for

anal

ysis

• M

ay b

e lim

ited

to

info

rma-

tion

with

in li

ne o

f “si

ght”

of

cam

era

• W

ell-

fund

ed h

and

was

hing

pro

mot

ion

pro

-gr

am e

valu

atio

ns o

r re

sear

ch s

tud

ies

• S

amp

le s

ize

calc

ulat

ions

, and

con

sulta

tion

with

per

sons

with

ep

idem

iolo

gica

l and

sta

-tis

tical

exp

ertis

e, m

ay in

dic

ate

feas

ibili

ty o

f th

is a

pp

roac

h fo

r ev

alua

tion

of e

ven

smal

l ha

ndw

ashi

ng p

rom

otio

n p

rogr

ams

• M

ay b

e p

artic

ular

ly u

sefu

l for

faci

lity-

bas

ed

obse

rvat

ion

whe

re la

rge

num

ber

s of

eve

nts

mig

ht o

ccur

(e.g

., sc

hool

s, h

ealth

care

fa

cilit

ies)




Mea

sure

men

t

Str

ateg

yR

equi

rem

ents

fo

r U

seA

dva

ntag

esD

isad

vant

ages

Ap

pro

pri

ate

Set

ting

fo

r U

se

Sen

sor-

bas

ed

Met

hod

s•

Eq

uip

men

t, s

uch

as s

oap

s w

ith

acce

lero

met

ers

• Tr

aini

ng in

the

pre

par

a-tio

n, in

itial

izat

ion,

dep

loy-

men

t, d

ata

dow

nloa

d, a

nd

dat

a an

alys

is

• Is

ob

ject

ive

• D

oes

not

req

uire

the

pre

s-en

ce o

f a h

uman

ob

serv

er

• D

epen

din

g on

the

tec

hnol

-og

y, s

enso

r d

ata

may

not

re-

flect

ind

ivid

ual-

leve

l beh

avio

r•

May

or

may

not

pro

vid

e in

-fo

rmat

ion

on h

and

was

hing

at

criti

cal t

imes

• C

urre

ntly

ava

ilab

le t

echn

olo-

gies

req

uire

hig

hly

trai

ned

st

aff

• A

re c

ostly

in t

erm

s of

eq

uip

-m

ent

need

s•

May

not

be

usef

ul in

set

tings

in

whi

ch b

ar s

oap

is u

sed

fo

r m

ultip

le p

urp

oses

, or

in

whi

ch li

qui

d o

r p

owd

er s

oap

is

com

mon

ly u

sed

• W

ell-

fund

ed h

and

was

hing

pro

mot

ion

pro

-gr

am e

valu

atio

ns o

r re

sear

ch s

tud

ies

• S

amp

le s

ize

calc

ulat

ions

, and

con

sulta

tion

with

per

sons

with

ep

idem

iolo

gica

l and

sta

-tis

tical

exp

ertis

e, m

ay in

dic

ate

feas

ibili

ty o

f th

is a

pp

roac

h fo

r ev

alua

tion

of e

ven

smal

l ha

ndw

ashi

ng p

rom

otio

n p

rogr

ams

TAB

LE

1:

SU

MM

AR

Y O

F S

TR

ATE

GIE

S T

O M

EA

SU

RE

HA

ND

WA

SH

ING

BE

HA

VIO

R (C

ont

inue

d)


www.wsp.org 23

Here, we provide recommendations for the measurement of handwashing behavior in low- and middle-income coun-try contexts. The recommendations below are made on the basis of ease of data collection and potential cost to the pro-gram or study. The focus here is on the measurement of handwashing behavior, although we comment on the util-ity of collecting data regarding knowledge, attitudes, and health outcomes in some of these contexts.

Well-funded Handwashing Promotion Programs or Research StudiesWell-funded programs or studies have the resources to in-volve experienced researchers with research and statistical expertise. Ideally, such studies should strive to use the most rigorous methods to measure handwashing behavior. Spe-cifically, structured observations are recommended. Struc-tured observation data can elucidate handwashing behavior for specific household members, e.g., primary caregivers of young children, and/or during particular critical times, e.g., after defecation.

Rapid observations, which are proxies, should also be in-cluded among measures of handwashing behavior used in well-funded programs. These observations provide useful information on the facilitating environment found in the home for good handwashing behavior.

For measurement of changes in knowledge or attitudes, or exposure to handwashing promotion programs or specific messages, questionnaires may prove useful. As noted above, the use of questionnaires for measurement of handwashing behavior is not recommended, since self-reported hand-washing behavior overestimates observed behavior.

At present, random or critical-time measurement of hand contamination is also not recommended as a measure of handwashing behavior, given the substantial variability de-tected in several studies described above. But, as detailed below, well-funded programs or research studies may serve as opportunities for improving upon this measure. Future studies should address the utility of indicator organisms other than E. coli, whether variability in hand contamina-tion is evident in other laboratories, and the relationship

between hand contamination and health outcomes. Well-funded studies may also be ideal for testing of new technol-ogies, such as sensors, for the measurement of handwashing behavior.

Several questions of import may be answered in the con-text of well-funded public health program evaluations and research studies:

• There is a fundamental gap in the literature on the relationship between the various measures of handwashing behavior and health outcomes. For most measures, there is a paucity of information on whether changes in a given handwashing measure are correlated with changes in risk of health outcomes of interest, such as diarrhea and respiratory infections. We strongly recommend that well-funded research studies and programs include measurement of both behavioral outcomes and health outcomes in the same study populations, and preferably in a longitu-dinal fashion, in order to examine these relationships in detail.

• Well-funded programs and research studies may be opportunities to validate novel methods, such as video observation, sensor-based methods, and hand cleansing product consumption.

• The utility of handwashing indices may be further examined, particularly in relationship to health outcomes, and as a means of identifying important explanatory factors associated with handwashing behavior.

• At present, there is still a paucity of published effec-tiveness data regarding the impact of public health programs on behavioral and health outcomes. It is strongly recommended that effectiveness data (posi-tive, negative, and neutral) be published in peer-reviewed literature in order to inform the public health community, policy makers, and funding agencies.

Handwashing Promotion Programs with Minimal FundingIdeally, these programs, as better-funded programs, would obtain objective measurement of handwashing behavior

Recommendations for Various ScenariosIV.


Practical Guidance for Measuring Handwashing Behavior: 2013 Update Recommendations for Various Scenarios


with structured observations. Cost is the primary limiting factor. Program evaluation staff are strongly encouraged to consult statistical and/or epidemiologic expertise in order to determine required sample sizes for measurement of handwashing behavior using structured observations, or sensor-based methods. Indeed, these more robust but more costly methods may be carried out in a subset of partici-pants. Universal recommendations regarding sample sizes cannot be made here, given the diversity in program types, evaluation designs, and program goals and targets.

Rapid observations, which are proxies, are certainly recom-mended as efficient measures of handwashing behavior in meagerly-funded public health programs. Also, such pro-grams may consider other approaches, perhaps in a subset of sample populations; these include structured observa-tion, video observation, inspections of hand cleanliness, and hand cleansing product consumption.

Questionnaires remain useful for measurement of knowl-edge, attitudes, and program exposure.

Nationally Representative Surveys, e.g., DHS or MICS SurveysThe Demographic and Health Surveys are described as “nationally-representative household surveys that provide data for a wide range of monitoring and impact evaluation indicators in the areas of population, health, and nutrition.” The multi-indicator cluster surveys (MICS) are conducted by UNICEF and may be described similarly. DHS and MICS surveys are conducted every three to five years in most low- and middle-income countries. Handwashing is only one of a myriad number of topics covered in these surveys and, thus, measurement of handwashing behavior is necessarily restricted to the most efficiently administered questions. It is not feasible to do more intensive measure-ments, such as structured observations, in the context of these large nationally representative surveys. Therefore, the use of rapid observations of handwashing materials is rec-ommended and has been adopted.


Practical Guidance for Measuring Handwashing Behavior: 2013 Update References

www.wsp.org 25

Aiello, A.E., R.M. Coulborn, V. Perez, E.L. Larson. 2008. “Effect of Hand Hygiene on Infectious Disease Risk in the Community Setting: A Meta-Analysis.” Ameri-can Journal of Public Health 98:1372–1381.

Armellino, D., E. Hussain, M.E. Schilling, et al. 2012. “Using High-Technology to Enforce Low-Technology Safety Measures: The Use of Third-Party Remote Video Auditing and Real-Time Feedback in Health-care.” Clinical Infectious Diseases 54:1–7.

Aunger, R., W.P. Schmidt, A. Ranpura, et al. 2010. “Three Kinds of Psychological Determinants for Hand-Washing Behaviour in Kenya.” Social Science & Medicine 70:383–91.

Bentley, M., M. Boot, J. Gittelsohn, R. Stallings. 1994. The Use of Structured Observations in the Study of Health Behaviour. The Hague, The Netherlands: IRC International Water and Sanitation Centre.

Biran, A., T. Rabie, W. Schmidt, S. Juvekar, S. Hirve, V. Curtis. 2008. “Comparing the Performance of Indicators of Hand-Washing Practices in Rural Indian Households.” Tropical Medicine & International Health 13:278–85.

Biran, A., W.P. Schmidt, R. Wright, et al. 2009. “The Effect of a Soap Promotion and Hygiene Education Campaign on Handwashing Behaviour in Rural India: A Cluster Randomised Trial.” Tropical Medicine & International Health 14:1303–14.

Burton, M., E. Cobb, P. Donachie, G. Judah, V. Curtis, W.P. Schmidt. 2011. “The Effect of Handwashing with Water or Soap on Bacterial Contamination of Hands.” International Journal of Environmental Research and Public Health 8:97–104.

Byrt, T., J. Bishop, J.B. Carlin. 1993. “Bias, Preva-lence and Kappa.” Journal of Clinical Epidemiology 46:423–9.

Cairncross, S., C. Hunt, S. Boisson, et al. 2010. “Water, Sanitation and Hygiene for the Prevention of Diarrhoea.” International Journal of Epidemiology 39(Suppl 1):i193–205.

Centers for Disease Control (CDC). 2002. “Guideline for Hand Hygiene in Health-Care Settings: Recommen-dations of the Healthcare Infection Control Practices

Advisory Committee and the HICPAC/SHEA/APIC/IDSA Hand Hygiene Task Force.” Morbidity and Mortality Weekly Report 51(RR-16):1–56.

Coombes, Yolande and Jacqueline Devine. 2010. “In-troducing FOAM: A Framework to Analyze Hand-washing Behaviors to Design Effective Handwashing Programs.” Working Paper, World Bank/Water and Sanitation Program.

Cousens, S., B. Kanki, S. Toure, I. Diallo, V. Curtis. 1996. “Reactivity and Repeatability of Hygiene Behaviour: Structured Observations from Burkina Faso.” Social Science & Medicine 43:1299–308.

Curtis,Valerie A., L.O. Danquah, R.V. Aunger. 2009. “Planned, Motivated and Habitual Hygiene Behav-iour: An Eleven Country Review.” Health Educa-tion Research. Published online 13 March 2009. doi:10.1093/her/cyp002

Curtis, Valerie. 2003. “Talking Dirty: How to Save a Mil-lion Lives.” International Journal of Environmental Health Research and Public Health 13(Suppl 1):S73–9.

Curtis, Valerie., B. Kanki, S. Cousens, et al. 2001. “Evi-dence of Behavior Change Following a Hygiene Pro-motion Programme in Burkina Faso.” Bulletin of the World Health Organization 79:518–27.

Curtis, Valerie and S. Cairncross. 2003. “Effect of Wash-ing Hands with Soap on Diarrhoea Risk in the Community: A Systematic Review.” Lancet Infectious Diseases 3:275–81.

Danquah, L. 2010. “Measuring Hand Washing Behaviour: Methodological and Validity Issues.” Presented at South Asia Hygiene Practitioners’ Workshop. Dhaka, Bangladesh.

Fagernes, M. and E. Lingaas. 2011. “Factors Interfering with the Microflora on Hands: A Regression Analysis of Samples from 465 Healthcare Workers.” Journal of Advanced Nursing 67:297–307.

Freeman, M.C., L.E. Greene, R. Dreibelbis, et al. 2011. “Assessing the Impact of a School-Based Water Treatment, Hygiene and Sanitation Programme on Pupil Absence in Nyanza Province, Kenya: A Cluster-Randomized Trial.” Tropical Medicine & International Health 17:380–391.

References




Judah, G., R. Aunger, W.P. Schmidt, S. Michie, S. Granger, V. Curtis. 2009. “Experimental Pretesting of Hand-Washing Interventions in a Natural Setting.” American Journal of Public Health 99(Suppl 2):S405–11.

Last, J.M., editor. 2001. A Dictionary of Epidemiology. 4th ed. New York: Oxford University Press.

Luby, S.P. and A.K. Halder. 2008. “Associations among Handwashing Indicators, Wealth, and Symptoms of Childhood Respiratory Illness in Urban Bangladesh.” Tropical Medicine & International Health 13:835–44.

Luby, S.P., A.K. Halder, C. Tronchet, S. Akhter, A. Bhuiya, R.B. Johnston. 2009. “Household Charac-teristics Associated with Handwashing with Soap in Rural Bangladesh.” American Journal of Tropical Medi-cine and Hygiene 81:882–7.

Luby, S.P., A.K. Halder, T. Huda, L. Unicomb, R.B. Johnston. 2011. “The Effect of Handwashing at Recommended Times with Water Alone and with Soap on Child Diarrhea in Rural Bangladesh: An Observational Study.” PLOS Medicine 8(6): e1001052. doi:10.1371/journal.pmed.1001052.

Luby, S.P., A.K. Halder, T.M. Huda, L. Unicomb, R.B. Johnston. 2011. “Using Child Health Outcomes to Identify Effective Measures of Handwashing.” American Journal of Tropical Medicine and Hygiene 85:882–92.

Luby, S.P., M. Agboatwalla, A. Bowen, E. Kenah, Y. Sharker, R.M. Hoekstra. 2009. “Difficulties in Main-taining Improved Handwashing Behavior, Karachi, Pakistan.” American Journal of Tropical Medicine and Hygiene 81:140–5.

Luby, S.P., M. Agboatwalla, A. Raza, et al. 2001. “Micro-biologic Effectiveness of Hand Washing with Soap in an Urban Squatter Settlement, Karachi, Pakistan.” Epidemiology and Infection 127:237–44.

Luby, S.P., M. Agboatwalla, W. Billhimer, R.M. Hoekstra. 2007. “Field Trial of a Low Cost Method to Evaluate Hand Cleanliness.” Tropical Medicine & International Health 12:765–71.

Luby, S.P., M.A. Kadir, M.A. Yushuf Sharker, F. Yeasmin, L. Unicomb, M. Sirajul Islam. 2010. “A Community-Randomised Controlled Trial Promoting Waterless Hand Sanitizer and Handwashing with Soap, Dhaka, Bangladesh.” Tropical Medicine & International Health 15:1508–16.

Gadgil, M.A., M.A.Y. Sharker, P.K. Ram, L. Unicomb, S.P. Luby. 2010. “Pilot Study of Serial Soap Weights as New Method of Measuring Handwashing in Dhaka, Bangladesh.” Presented at the annual meeting for the American Society of Tropical Medicine and Hygiene. Atlanta, Georgia, November 3–7.

Gadgil, M.A., M.A.Y. Sharker, L. Unicomb, S.P. Luby, P.K. Ram. 2011. “Consistent Soap Availability Cor-relates with Use of Inexpensive Soap Products and Improved Handwashing in Low-Income Households in Dhaka, Bangladesh.” Presented at the annual meet-ing for the American Society of Tropical Medicine and Hygiene, Philadelphia, Pennsylvania, December 4–8.

Grayson, M.L., S. Melvani, J. Druce, et al. 2009. “Efficacy of Soap and Water and Alcohol-Based Hand-Rub Preparations against Live H1N1 Influenza Virus on the Hands of Human Volunteers.” Clinical Infectious Diseases 48:285–91.

Halder, A.K., C. Tronchet, S. Akhter, A. Bhuiya, R. John-ston, S.P. Luby. 2010. “Observed Hand Cleanliness and Other Measures of Handwashing Behavior in Rural Bangladesh.” BMC Public Health 10:545.

Hernandez, O., J. Devine, J. Karver, C. Chase, Y. Coombes. 2012. “Measuring the Behavioral Deter-minants of Handwashing with Soap.” Working Paper, World Bank/Water and Sanitation Program.

Hoey, J., P. Poupart, A. von Bertoldi, T. Craig, C. Boutil-ier, A. Mihailidis. 2010. “Automated Handwashing Assistance for Persons with Dementia Using Video and a Partially Observable Markov Decision Pro-cess.” Computer Vision and Image Understanding 114:503–19.

Hoque, B.A., D. Mahalanabis, M.J. Alam, M.S. Islam. 1995. “Post-Defecation Handwashing in Bangladesh: Practice and Efficiency Perspectives.” Public Health 109:15–24.

Huda, T.M., L. Unicomb, R.B. Johnston, A.K. Halder, M.A. Yushuf Sharker, S.P. Luby. 2012. “Interim Evaluation of a Large Scale Sanitation, Hygiene and Water Improvement Programme on Childhood Diarrhea and Respiratory Disease in Rural Bangla-desh.” Social Science & Medicine 75:604–611. doi: 10.1016/j.socscimed.2011.10.042.

International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B). 2008.“Handwashing Behav-ior in Rural Bangladesh.” Health and Science Bulletin Vol. 6, No. 3.



www.wsp.org 27

Pickering, A.J., T.R. Julian, S. Mamuya, A.B. Boehm, J. Davis. 2011. “Bacterial Hand Contamination among Tanzanian Mothers Varies Temporally and Following Household Activities.” Tropical Medicine & Interna-tional Health 16:233–9.

Pinfold, J.V. 1990. “Faecal Contamination of Water and Fingertip-Rinses as a Method for Evaluating the Effect of Low-Cost Water Supply and Sanitation Activities on Faeco-Oral Disease Transmission. I. A Case Study in Rural North-East Thailand.” Epidemiology and In-fection 105:363–75.

Pinfold, J.V., N.J. Horan. 1996. “Measuring the Effect of a Hygiene Behaviour Intervention by Indicators of Behaviour and Diarrhoeal Disease.” Transactions of the Royal Society of Tropical Medicine and Hygiene 90:366–71.

Rabie, Tamer and Valerie Curtis. 2006. “Handwashing and Risk of Respiratory Infections: A Quantitative Systematic Review.” Tropical Medicine & International Health 11:258–67.

Ram, P., S. Luby, A. Halder, M. Islam, S. Granger. 2010. “Improving Measures of Handwashing Behavior.” Working Paper, World Bank/Water and Sanitation Program.

Ram, P.K., A.K. Halder, S.P. Granger, et al. 2010. “Is Structured Observation a Valid Technique to Measure Handwashing Behavior? Use of Acceleration Sensors Embedded in Soap to Assess Reactivity to Structured Observation.” American Journal of Tropical Medicine and Hygiene 83:1070–6.

Ram, P.K., I. Jahid, A.K. Halder, et al. 2011. “Variability in Hand Contamination Based on Serial Measure-ments: Implications for Assessment of Hand-Cleansing Behavior and Disease Risk.” American Journal of Tropical Medicine and Hygiene 84:510–6.

Rhee, V., L.C. Mullany, S.K. Khatry, et al. 2008. “Mater-nal and Birth Attendant Hand Washing and Neonatal Mortality in Southern Nepal.” Archives of Pediatric & Adolescent Medicine 162:603–8.

Sagerman, D., F.A. Nizame, M. Nuruzzaman, J. Yu, S.P. Luby, P.K. Ram. 2010. “Impact of Complexity of Handwashing Instructions on Adherence in a Low Income Setting, Dhaka, Bangladesh.” Presented at the annual meeting for the American Society of Tropical Medicine and Hygiene. Philadelphia, Pennsylvania, December 4–8.

Luby, Stephen P., M. Agboatwalla, D.R. Feikin, et al. 2005. “Effect of Handwashing on Child Health: A Randomised Controlled Trial.” Lancet 366:225–33.

Luby, Stephen, M. Agboatwalla, J. Painter, A. Altaf, W. Billhimer, R. Hoekstra. 2004. “Effect of Intensive Handwashing Promotion on Childhood Diarrhea in High-Risk Communities in Pakistan: A Randomized Controlled Trial.” JAMA 291:2547–54.

Manun’Ebo, M., S. Cousens, P. Haggerty, M. Kalengaie, A. Ashworth, B. Kirkwood. 1997. “Measuring Hy-giene Practices: A Comparison of Questionnaires with Direct Observations in Rural Zaire.” Tropical Medicine & International Health 2:1015–21.

McAteer, J., S. Stone, C. Fuller, et al. 2008. “Development of an Observational Measure of Healthcare Worker Hand-Hygiene Behaviour: The Hand-Hygiene Ob-servation Tool (HHOT).” Journal of Hospital Infection 68:222–9.

McGuckin, M., R. Waterman, J. Govednik. 2009. “Hand Hygiene Compliance Rates in the United States—A One-Year Multicenter Collaboration Using Product/Volume Usage Measurement and Feedback.” American Journal of Medical Quality 24:205–13.

Munro, A.D., R.G. Munro, inventors. 2009. “RF Con-trolled Devices to Increase Compliance with Hand-washing Protocols.” Patent Number 20090237254, United States of America.

Pickering, A.J., A.B. Boehm, M. Mwanjali, J. Davis. 2010. “Efficacy of Waterless Hand Hygiene Compared with Handwashing with Soap: A Field Study in Dar es Sa-laam, Tanzania.” American Journal of Tropical Medicine and Hygiene 82:270–8.

Pickering, A.J., J. Davis, A. Blum, et al. 2011. “Access to Waterless Hand Sanitizer Improves Hand Cleaning Behavior and Health at Primary Schools in Kibera, Kenya.” Presented at the annual meeting for the American Society of Tropical Medicine and Hygiene. Philadelphia, Pennsylvania, December 4–8.

Pickering, A.J., J. Davis, A.B. Boehm. 2011. “Efficacy of Alcohol-Based Hand Sanitizer on Hands Soiled with Dirt and Cooking Oil.” Journal of Water and Health 9:429–33.

Pickering, A.J., J. Davis, S.P. Walters, et al. 2010. “Hands, Water, and Health: Fecal Contamination in Tanza-nian Communities with Improved, Non-Networked Water Supplies.” Environmental Science & Technology 44:3267–72.




Talaat, M., S. Afifi, E. Dueger, et al. 2011. “Effects of Hand Hygiene Campaigns on Incidence of Laboratory-Confirmed Influenza and Absenteeism in Schoolchildren, Cairo, Egypt.” Emerging Infectious Diseases 17:619–25.

Unicomb, L.E., T.M.N. Huda, R.B. Johnston, et al. 2010. “The Relationship between Diarrheal Death and Measures of Water, Sanitation, and Hygiene among Children < 5 Years of Age in Bangladesh.” Poster pre-sented at the annual meeting for the American Society of Tropical Medicine and Hygiene, Atlanta, Georgia, November 3–7.

Verplanken, B. and S. Orbell. 2003. “Reflections on Past Behavior: A Self-Report Index of Habit Strength.” Journal of Applied Social Psychology 33:1313–30.

Vindigni, S.M., P.L. Riley, M. Jhung. 2011. “Systematic Review: Handwashing Behaviour in Low- to Middle-Income Countries: Outcome Measures and Behaviour Maintenance.” Tropical Medicine & International Health 16:466–77.

Wang, A., L. McMahan, M. Sobsey, C. Stauber, S. Rut-stein. 2011. “Evaluation of Household Microbial Water Quality Testing in a Peruvian Demographic and Health Pilot Study Using the Portable Compart-ment Bag Test (CB) for E. coli.” Presented at the an-nual meeting for the American Society of Tropical Medicine and Hygiene. Philadelphia, Pennsylvania, December 4–8.

Webb, A.L., A.D. Stein, U. Ramakrishnan, V.S. Hertz-berg, M. Urizar, R. Martorell. 2006. “A Simple Index to Measure Hygiene Behaviours.” International Journal of Epidemiology 35:1469–77.

Whaley, L. and J. Webster. 2011. “The Effectiveness and Sustainability of Two Demand-Driven Sanitation and Hygiene Approaches in Zimbabwe.” Journal of Water, Sanitation, and Hygiene for Development 1:20–36.

Yalcin, S.S., S. Yalcin, S. Altin. 2004. “Hand Washing and Adolescents: A Study from Seven Schools in Konya, Turkey.” International Journal of Adolescent Medicine and Health 16:371–6.

Sandora, T.J., E.M. Taveras, M.C. Shih, et al. 2005. “A Randomized, Controlled Trial of a Multifaceted Inter-vention Including Alcohol-Based Hand Sanitizer and Hand-Hygiene Education to Reduce Illness Transmis-sion in the Home.” Pediatrics 116:587–94.

Schmidt, W.P., R. Aunger, Y. Coombes, et al. 2009. “De-terminants of Handwashing Practices in Kenya: The Role of Media Exposure, Poverty and Infrastructure.” Tropical Medicine & International Health 14:1534–41.

Silk, B.J., S. Doshi, D. Dutt, et al. 2010. “Hand Hygiene and Radiographically-Confirmed Pneumonia among Young Children Living in Urban Dhaka, Bangladesh: Preliminary Results from a Case-Control Study.” Poster presented at the annual meeting for the Ameri-can Society of Tropical Medicine and Hygiene. Atlanta, Georgia, November 3–7.

Sim, J., C.C. Wright. 2005. “The Kappa Statistic in Reli-ability Studies: Use, Interpretation, and Sample Size Requirements.” Physical Therapy 85:257–68.

Simmerman, J.M., P. Suntarattiwong, J. Levy, et al. 2011. “Findings from a Household Randomized Controlled Trial of Hand Washing and Face Masks to Reduce In-fluenza Transmission in Bangkok, Thailand.” Influenza and Other Respiratory Viruses 5:256–67.

Stanton, B.F., J.D. Clemens, K.M. Aziz, M. Rahman. 1987. “Twenty-Four-Hour Recall, Knowledge-Atti-tude-Practice Questionnaires, and Direct Observa-tions of Sanitary Practices: A Comparative Study.” Bulletin of the World Health Organization 65:217–22.

Stevenson, R.J., T.I. Case, D. Hodgson, R. Porzig-Drummond, J. Barouei, M.J. Oaten. 2009. “A Scale for Measuring Hygiene Behavior: Development, Reliability and Validity.” American Journal of Infection Control 37:557–64.

Strina, A., S. Cairncross, M.L. Barreto, C. Larrea, M.S. Prado. 2003. “Childhood Diarrhea and Observed Hy-giene Behavior in Salvador, Brazil.” American Journal of Epidemiology 157:1032–8.



8714-Book.indd 308714-Book.indd 30 2/14/13 2:18 PM2/14/13 2:18 PM

Date post:	17-Apr-2018
Category:	Documents
Upload:	lamtruc
View:	220 times
Download:	5 times

Practical Guidance for Measuring Handwashing … Guidance for Measuring Handwashing Behavior: 2013...

Documents